This document shows you how to select specific compute classes to run workloads that
have unique hardware requirements in your Google Kubernetes Engine (GKE)
Autopilot clusters. Before reading this document,
ensure that you're familiar with the concept
of compute classes in GKE Autopilot.
Overview of Autopilot compute classes
Autopilot offers compute classes that are designed to run
workloads that have specific hardware requirements. These compute classes are
useful for workloads such as machine learning and AI tasks, or running real-time
high traffic databases.
These compute classes are a subset of the Compute Engine
machine series, and offer
flexibility beyond the default Autopilot general-purpose compute class.
For example, the Scale-Out class turns off simultaneous multi-threading so that each
vCPU is one physical core.
Based on your individual Pod needs, you can configure your regular
Autopilot Pods or your Spot Pods to request nodes backed by
these compute classes. You can also request specific CPU architecture, such as
Arm, in compute classes that
support that architecture.
Before you begin
Before you start, make sure that you have performed the following tasks:
If you want to use the Google Cloud CLI for this task,
install and then
initialize the
gcloud CLI. If you previously installed the gcloud CLI, get the latest
version by running gcloud components update.
To tell Autopilot to place your Pods on a specific compute class,
specify the cloud.google.com/compute-class label in a
nodeSelector
or a node affinity rule,
such as in the following examples:
You can also request specific compute classes for your Spot Pods.
Specify resource requests
When you choose a compute class, make sure that you specify resource requests
for your Pods based on the
Minimum and maximum resource requests
for your selected class. If your requests are less than the minimum,
Autopilot automatically scales your requests up. However, if your
requests are greater than the maximum, Autopilot does not deploy your
Pods and displays an error message.
Choose a CPU architecture
Some compute classes support multiple CPU architectures. For example, the
Scale-Out class supports both Arm and x86 architectures. If you
don't request a specific architecture, Autopilot provisions nodes that
have the default architecture of the specified compute class. If your Pods need
to use a different architecture, request that architecture in your node selector
or node affinity rule, alongside your compute class request. The compute class
that you request must support the CPU architecture you specify.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-08-26 UTC."],[],[],null,["# Choose compute classes for Autopilot Pods\n\n[Autopilot](/kubernetes-engine/docs/concepts/autopilot-overview)\n\n*** ** * ** ***\n\nThis document shows you how to select specific compute classes to run workloads that\nhave unique hardware requirements in your Google Kubernetes Engine (GKE)\nAutopilot clusters. Before reading this document,\nensure that you're familiar with the concept\nof [compute classes in GKE Autopilot](/kubernetes-engine/docs/concepts/autopilot-compute-classes).\n\nOverview of Autopilot compute classes\n-------------------------------------\n\nAutopilot offers *compute classes* that are designed to run\nworkloads that have specific hardware requirements. These compute classes are\nuseful for workloads such as machine learning and AI tasks, or running real-time\nhigh traffic databases.\n\nThese compute classes are a subset of the Compute Engine\n[machine series](/compute/docs/machine-types#machine_type_comparison), and offer\nflexibility beyond the default Autopilot general-purpose compute class.\nFor example, the `Scale-Out` class turns off simultaneous multi-threading so that each\nvCPU is one physical core.\n\nBased on your individual Pod needs, you can configure your regular\nAutopilot Pods or your Spot Pods to request nodes backed by\nthese compute classes. You can also request specific CPU architecture, such as\n[Arm](https://www.arm.com/architecture), in compute classes that\nsupport that architecture.\n\nBefore you begin\n----------------\n\nBefore you start, make sure that you have performed the following tasks:\n\n- Enable the Google Kubernetes Engine API.\n[Enable Google Kubernetes Engine API](https://console.cloud.google.com/flows/enableapi?apiid=container.googleapis.com)\n- If you want to use the Google Cloud CLI for this task, [install](/sdk/docs/install) and then [initialize](/sdk/docs/initializing) the gcloud CLI. If you previously installed the gcloud CLI, get the latest version by running `gcloud components update`. **Note:** For existing gcloud CLI installations, make sure to set the `compute/region` [property](/sdk/docs/properties#setting_properties). If you use primarily zonal clusters, set the `compute/zone` instead. By setting a default location, you can avoid errors in the gcloud CLI like the following: `One of [--zone, --region] must be supplied: Please specify location`. You might need to specify the location in certain commands if the location of your cluster differs from the default that you set.\n\n\u003c!-- --\u003e\n\n- [Ensure that you have a GKE Autopilot cluster](/kubernetes-engine/docs/how-to/creating-an-autopilot-cluster) running GKE version 1.24.1-gke.1400 or later.\n\nRequest a compute class in your Autopilot Pod\n---------------------------------------------\n\nTo tell Autopilot to place your Pods on a specific compute class, specify the `cloud.google.com/compute-class` label in a [nodeSelector](https://kubernetes.io/docs/concepts/scheduling-eviction/assign-pod-node/#nodeselector) or a [node affinity rule](https://kubernetes.io/docs/concepts/scheduling-eviction/assign-pod-node/#node-affinity), such as in the following examples:\n\n### nodeSelector\n\n```yaml\n apiVersion: apps/v1\n kind: Deployment\n metadata:\n name: hello-app\n spec:\n replicas: 3\n selector:\n matchLabels:\n app: hello-app\n template:\n metadata:\n labels:\n app: hello-app\n spec:\n nodeSelector:\n cloud.google.com/compute-class: \"\u003cvar translate=\"no\"\u003eCOMPUTE_CLASS\u003c/var\u003e\"\n containers:\n - name: hello-app\n image: us-docker.pkg.dev/google-samples/containers/gke/hello-app:1.0\n resources:\n requests:\n cpu: \"2000m\"\n memory: \"2Gi\"\n \n```\n\nReplace \u003cvar translate=\"no\"\u003eCOMPUTE_CLASS\u003c/var\u003e with the name of the [compute class](/kubernetes-engine/docs/concepts/autopilot-compute-classes#when-to-use)\nbased on your use case, such as `Scale-Out`. If you select `Accelerator`,\nyou must also specify a compatible GPU. For instructions, see [Deploy GPU workloads in Autopilot](/kubernetes-engine/docs/how-to/autopilot-gpus). If\nyou select `Performance`, you can optionally select a Compute Engine machine\nseries in the node selector. If you don't specify a machine series, GKE uses the\nC4 machine series depending on [regional\navailability](/compute/docs/regions-zones#available). For instructions, see [Run CPU-intensive workloads with optimal\nperformance](/kubernetes-engine/docs/how-to/performance-pods).\n\n### nodeAffinity\n\n```yaml\n apiVersion: apps/v1\n kind: Deployment\n metadata:\n name: hello-app\n spec:\n replicas: 3\n selector:\n matchLabels:\n app: hello-app\n template:\n metadata:\n labels:\n app: hello-app\n spec:\n terminationGracePeriodSeconds: 25\n containers:\n - name: hello-app\n image: us-docker.pkg.dev/google-samples/containers/gke/hello-app:1.0\n resources:\n requests:\n cpu: \"2000m\"\n memory: \"2Gi\"\n ephemeral-storage: \"1Gi\"\n affinity:\n nodeAffinity:\n requiredDuringSchedulingIgnoredDuringExecution:\n nodeSelectorTerms:\n - matchExpressions:\n - key: cloud.google.com/compute-class\n operator: In\n values:\n - \"\u003cvar translate=\"no\"\u003eCOMPUTE_CLASS\u003c/var\u003e\"\n \n```\n\nReplace \u003cvar translate=\"no\"\u003eCOMPUTE_CLASS\u003c/var\u003e with the name of the\n[compute\nclass](/kubernetes-engine/docs/concepts/autopilot-compute-classes#when-to-use) based on your use case, such as `Scale-Out`. If you select\n`Accelerator`, you must also specify a compatible GPU. For instructions, see [Deploy GPU workloads in Autopilot](/kubernetes-engine/docs/how-to/autopilot-gpus).\nIf you select `Performance`, you can optionally select a\nCompute Engine machine series in the node selector. If you don't specify a machine series, GKE uses the C4\nmachine series depending on [regional\navailability](/compute/docs/regions-zones#available). For instructions, see [Run CPU-intensive workloads with\noptimal performance](/kubernetes-engine/docs/how-to/performance-pods).\n\nYou can also request specific compute classes for your Spot Pods.\n\n### Specify resource requests\n\nWhen you choose a compute class, make sure that you specify resource requests\nfor your Pods based on the\n[Minimum and maximum resource requests](/kubernetes-engine/docs/concepts/autopilot-resource-requests#min-max-requests)\nfor your selected class. If your requests are less than the minimum,\nAutopilot automatically scales your requests up. However, if your\nrequests are greater than the maximum, Autopilot does not deploy your\nPods and displays an error message.\n\nChoose a CPU architecture\n-------------------------\n\nSome compute classes support multiple CPU architectures. For example, the\n`Scale-Out` class supports both Arm and x86 architectures. If you\ndon't request a specific architecture, Autopilot provisions nodes that\nhave the default architecture of the specified compute class. If your Pods need\nto use a different architecture, request that architecture in your node selector\nor node affinity rule, alongside your compute class request. The compute class\nthat you request must support the CPU architecture you specify.\n\nFor instructions, refer to\n[Deploy Autopilot Pods on Arm architecture](/kubernetes-engine/docs/how-to/autopilot-arm-workloads).\n\nWhat's next\n-----------\n\n- [Learn more about Autopilot cluster architecture](/kubernetes-engine/docs/concepts/autopilot-architecture).\n- [Learn about the lifecycle of Pods](https://kubernetes.io/docs/concepts/workloads/pods/pod-lifecycle/).\n- [Learn about the available Autopilot compute classes](/kubernetes-engine/docs/concepts/autopilot-compute-classes).\n- [Read about the default, minimum, and maximum resource requests for each\n platform](/kubernetes-engine/docs/concepts/autopilot-resource-requests)."]]